The budding of HIV and other retroviruses is a highly orchestrated process involving both viral and host proteins. Previous work from our laboratory showed that HIV budding occurs at specialized regions of the membrane known as lipid rafts. These membrane microdomains, highly enriched in cholesterol and sphingolipids, serve as platforms for lattices for signaling proteins. We and others have shown that these membrane domains also serve as sites for entry of HIV into cells. It is now well appreciated that cholesterol is critical for both virus budding and release. We and others have obtained evidence showing that HIV assembly and budding occurs in late endosomal compartments corresponding to multivesicular bodies.(MVBs). Evidence suggests that in T cells budding at the plasma membrane may also occur at membrane sites representing translocated late endosomal membrane. The central hypothesis of this proposal is that HIV particles are Trojan exosomes. Exosomes are virus-sized extracellular organelles produced in and released from MVBs. This model has important implications for HIV transmission, pathogenesis and vaccines. The model predicts that HIV biogenesis requires proteins involved in vesicular trafficking including the Rab family of GTPases and the NPC1 protein, which plays a critical role in cellular cholesterol homeostasis and late endosomal trafficking. We have obtained preliminary evidence supporting this idea. The goal of the proposed studies is to probe the role of these molecules in HIV infection. The specific aims are: 1) To study the role of NPCl and Rab proteins in HIV-1 infection. Studies will be conducted to determine the association of these proteins with virus particles and whether their cellular localization is altered by HIV-1 infection, siRNA will be used to turn off expression of these proteins and the effect on HIV replication determined. 2) To induce the Niemann-Pick phenotype in cells and determine effects on HIV.1 infection. The hydrophobic amine u18666A will be used to induce the Niemann Pick phenotype in which the trafficking of late endosomes is impaired. The effect on HIV-1 infection will be studied. The results of these studies should provide important new insights into the biology of HIV and provide support for the Trojan exosome model. If correct this model holds the promise of providing new targets for anti-retroviral drugs and new approaches for prophylactic vaccines.